CN101432610B - Method and apparatus for lossless encoding of a source signal using a lossy encoded data stream and a lossless extension data stream - Google Patents
Method and apparatus for lossless encoding of a source signal using a lossy encoded data stream and a lossless extension data stream Download PDFInfo
- Publication number
- CN101432610B CN101432610B CN2007800156126A CN200780015612A CN101432610B CN 101432610 B CN101432610 B CN 101432610B CN 2007800156126 A CN2007800156126 A CN 2007800156126A CN 200780015612 A CN200780015612 A CN 200780015612A CN 101432610 B CN101432610 B CN 101432610B
- Authority
- CN
- China
- Prior art keywords
- signal
- coding
- data
- frequency spectrum
- decoding
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims description 31
- 238000001228 spectrum Methods 0.000 claims description 101
- 238000011002 quantification Methods 0.000 claims description 22
- 238000004458 analytical method Methods 0.000 claims description 21
- 230000003467 diminishing effect Effects 0.000 claims description 17
- 230000006978 adaptation Effects 0.000 claims description 11
- 230000003292 diminished effect Effects 0.000 claims description 11
- 230000005236 sound signal Effects 0.000 abstract description 8
- 230000002708 enhancing effect Effects 0.000 abstract 1
- 238000012856 packing Methods 0.000 description 22
- 239000002131 composite material Substances 0.000 description 13
- 230000003044 adaptive effect Effects 0.000 description 8
- 238000013139 quantization Methods 0.000 description 8
- 238000005070 sampling Methods 0.000 description 6
- 230000005540 biological transmission Effects 0.000 description 5
- 238000006243 chemical reaction Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000001360 synchronised effect Effects 0.000 description 3
- 230000008034 disappearance Effects 0.000 description 2
- 230000000873 masking effect Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000003556 assay Methods 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000003139 buffering effect Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000007429 general method Methods 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000007493 shaping process Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M7/00—Conversion of a code where information is represented by a given sequence or number of digits to a code where the same, similar or subset of information is represented by a different sequence or number of digits
- H03M7/30—Compression; Expansion; Suppression of unnecessary data, e.g. redundancy reduction
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L19/00—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis
- G10L19/0017—Lossless audio signal coding; Perfect reconstruction of coded audio signal by transmission of coding error
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L19/00—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis
- G10L19/04—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis using predictive techniques
- G10L19/16—Vocoder architecture
- G10L19/18—Vocoders using multiple modes
- G10L19/24—Variable rate codecs, e.g. for generating different qualities using a scalable representation such as hierarchical encoding or layered encoding
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L19/00—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L19/00—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis
- G10L19/02—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis using spectral analysis, e.g. transform vocoders or subband vocoders
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Theoretical Computer Science (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
- Transmission Systems Not Characterized By The Medium Used For Transmission (AREA)
Abstract
In lossy based lossless coding a PCM audio signal passes through a lossy encoder (41) to a lossy decoder (42). The lossy encoder provides a lossy bit stream. The difference signal (S<Diff>) between the PCM signal and the lossy decoder output is lossless encoded (52), providing an extension bit stream. The invention facilitates enhancing a lossy percep tual audio encoding/decoding by an extension that enables mathematically exact reproduction of the original waveform using enhanced de-correlation, and provides additional data for reconstructing at decoder site an intermediate-quality audio signal. The lossless extension can be used to extend the widely used mp3 encoding/decoding to lossless encoding/decoding and superior quality mp3 encoding/decoding.
Description
Technical field
The present invention relates to use the data stream of lossy coding (encode) and lossless extension data to flow method and the equipment that source signal is carried out lossless coding, the data stream of wherein said lossy coding and lossless extension data stream form the data stream of the lossless coding of described source signal together.By can be on mathematics accurately the growth data of (can't harm) reproduction original audio signal waveform strengthened and diminished sensing audio encoding (coding) data.
Background technology
The ultimate principle of lossless audio coding has been described in Fig. 1.The digital pcm sampled audio signal is not independent mutually.Signal decorrelation 11 is used for reducing this dependence before entropy coding 12.This process need is reversible, can recover original signal.Existing de-correlation technique is used linear prediction filtering (also being called linear predictive coding LPC), integer bank of filters (integerfilter-bank) and based on the mode that diminishes.
Ultimate principle based on the lossless coding that diminishes has been described in Fig. 2 and Fig. 3.Coded portion in Fig. 2 (left side), pcm audio input signal S
PCMTo diminishing demoder 22, and diminish demoder 25 to decoded portion (right side) by lossy encoder 21 as lossy bit stream.Lossy coding and decoding are used for signal is carried out decorrelation.In subtracter 23 with the output signal of demoder 22 from input signal S
PCMRemove, and the difference signal that obtains arrives non-damage decoder 27 by lossy encoder 24 as spread bit stream.The 26 output signal combinations with demoder 25 and 27 are to obtain original signal S again
PCMIn EP-B-0756386 and US-B-6498811 at audio coding this ultimate principle is disclosed, also at P.Craven, M.Gerzon, " Lossless Coding for Audio Discs ", J.Audio Eng.Soc., Vol.44, No.9, among the September1996 and at J.Koller, Th.Sporer, K.H.Brandenburg, " Robust Coding of High Quality Audio Signals ", AES103rd Convention, Preprint4621 has discussed this ultimate principle among the August1997.
In the lossy encoder in Fig. 3, pcm audio input signal S
PCMQuantification 32 by the sampling of analysis filterbank 31 and subband (sub-band) to coding and bit stream (packing) 33 that pack.By received signal S
PCMAnd control quantification from the sensor model counter 34 that analysis filterbank 31 receives corresponding informations.
At decoder-side, the lossy bit stream of decoding enters and is used for device 35 that bit stream is unpacked, enters then to be used for device 36 and composite filter group 37 that sub-band sample is decoded, diminishes PCM signal S behind the output decoder
Dec
In standard ISO/IEC11172-3 (MPEG-1Audio), describe example in detail at lossy coding and decoding.
Because lossy encoder is created in the frequency domain and shelters the proportional error signal S of (masking) threshold value
DiffSo not well with the signal decorrelation, thereby signal is suboptimum for entropy coding.Therefore, following discloses are paid close attention to error signal S
DiffSpecial processing.General method is that the variant with LPC decorrelation scheme is applied to error signal S
Diff: WO-A-9953677, US-A-20040044520, WO-A-2005098823.In EP-A-0905918, utilize feedback control loop with error signal S
DiffAmplitude be used for the quantization step of lossy encoder part, thereby produce error signal S with the quantification that is controlled in the lossy encoder
DiffBetter decorrelation.
Summary of the invention
When the lossless coding expansion was provided for lossless coding, expectation made this process become easy in the mode of scalable (scalable).
The problem that the present invention will solve is, is that lossy coding/decoding provides a kind of improved lossless coding/decoding expansion in scalable mode, and for example, lossy coding/decoding is based on mp3's (MPEG-1Audio Layer3).Utilize the coding/decoding method in disclosed coding method in the claim 1 and claim 3 and 5 to solve this problem.In claim 2,4 and 6 equipment that uses these methods is disclosed respectively.
The expansion of (lossless coding/decoding just) original waveform can be accurately reproduced in utilization of the present invention on mathematics, be convenient to strengthen diminishing sensing audio encoding/decoding.By embedding the frequency spectrum decorrelation in lossy encoder-demoder and embedding additional period LPC decorrelation, use the decorrelation that strengthens based on the lossless coding that diminishes, wherein do not need to transmit the LPC filter parameter.
Advantageously, lossless extension of the present invention can be used for widely used mp3 coding/decoding is expanded to lossless coding/decoding.
Advantageously, coding method of the present invention is fit to use the data stream of lossy coding and lossless extension data stream that source signal is carried out lossless coding, the data stream of described lossy coding and lossless extension data stream form the data stream of the lossless coding of described source signal together, and described method comprises step:
-described source signal is carried out lossy coding, wherein said lossy coding provides the data stream and the frequency spectrum albefaction data of described lossy coding;
-data of described lossy coding are diminished decoding accordingly, thus the signal of reconstruction standard decoding, and use described frequency spectrum albefaction data, construct the signal of high-quality decoding according to the signal of described standard decoding;
-form the difference signal between the signal that described source signal and described high-quality decode, and described difference signal is carried out lossless coding;
-difference signal of described coding is packed with described frequency spectrum albefaction data, to form described lossless extension data stream.
On the principle, encoding device of the present invention is suitable for using the data stream of lossy coding and lossless extension data stream that source signal is carried out lossless coding, the data stream of described lossy coding and lossless extension data stream form the data stream of the lossless coding of described source signal together, and described equipment comprises:
-being suitable for described source signal is carried out the device of lossy coding, wherein said lossy coding provides the data stream and the frequency spectrum albefaction data of described lossy coding;
-be suitable for the data of described lossy coding are diminished decoding accordingly, thus the signal of reconstruction standard decoding and use the signal of the data based described standard decoding of described frequency spectrum albefaction to construct the device of the signal of high-quality decoding;
-be suitable for forming described source signal between the signal of described high-quality decoding difference signal and be suitable for described difference signal is carried out lossless coding and be suitable for that described difference signal is carried out lossless coding packing to form the device of described lossless extension data stream with described frequency spectrum albefaction data.
On the principle, coding method of the present invention is suitable for the source signal data stream of lossless coding is decoded, described data stream is to obtain from the data stream of lossy coding and lossless extension data stream, the data stream of described lossy coding and lossless extension data stream form the data stream of the lossless coding of described source signal together
Wherein described source signal has been carried out lossy coding, described lossy coding provides the data stream and the frequency spectrum albefaction data of described lossy coding,
Wherein the data of described lossy coding have been carried out diminishing decoding accordingly, thereby the signal of reconstruction standard decoding and uses described frequency spectrum albefaction data, according to the signal configuration of described standard decoding the signal of high-quality decoding,
Wherein formed the difference signal between the signal of described source signal and described high-quality decoding, and described difference signal carried out lossless coding,
Wherein the difference signal of described lossless coding is packed with described frequency spectrum albefaction data and flows to form described lossless extension data,
Described method comprises step:
-described lossless extension data stream is unpacked, and the difference signal of described lossless coding is decoded, so that described difference signal and described frequency spectrum albefaction data to be provided;
-data stream of described lossy coding is diminished decoding, thus the signal of described standard decoding rebuild, and use described frequency spectrum albefaction data, rebuild the signal of described high-quality decoding according to the signal of described standard decoding;
-according to the difference signal of the lossless coding of described decoding and according to the signal of described high-quality decoding, form the source signal of rebuilding.
On the principle, encoding device of the present invention is suitable for the source signal data stream of lossless coding is decoded, described data stream is to obtain from the data stream of lossy coding and lossless extension data stream, the data stream of described lossy coding and lossless extension data stream form the data stream of the lossless coding of described source signal together
Wherein described source signal has been carried out lossy coding, described lossy coding provides the data stream and the frequency spectrum albefaction data of described lossy coding,
Wherein the data of described lossy coding have been carried out diminishing decoding accordingly, thereby have rebuild the signal of standard decoding, and used described frequency spectrum albefaction data, according to the signal configuration of described standard decoding the signal of high-quality decoding,
Wherein formed the difference signal between the signal of described source signal and described high-quality decoding, and described difference signal carried out lossless coding,
Wherein the difference signal of described lossless coding is packed with described frequency spectrum albefaction data and flows to form described lossless extension data,
Described equipment comprises:
-be suitable for described lossless extension data stream is unpacked and is suitable for the difference signal of described lossless coding is decoded so that the device of described difference signal and described frequency spectrum albefaction data to be provided;
-be suitable for the data stream of described lossy coding is diminished decoding, thus the signal of described standard decoding rebuild, and use described frequency spectrum albefaction data, rebuild the device of the signal of described high-quality decoding according to the signal of described standard decoding;
-be suitable for according to the lossless coding difference signal of described decoding and the device that forms the source signal of reconstruction according to the signal of described high-quality decoding.
On the principle, another kind of coding/decoding method of the present invention is suitable for the source signal data stream of lossless coding is decoded, described data stream is to obtain from the data stream of lossy coding and lossless extension data stream, the data stream of described lossy coding and lossless extension data stream form the data stream of the lossless coding of described source signal together
Wherein described source signal has been carried out lossy coding, described lossy coding provides the data stream and the frequency spectrum albefaction data of described lossy coding,
Wherein the data of described lossy coding have been carried out diminishing decoding accordingly, thereby have rebuild the signal of standard decoding, and used described frequency spectrum albefaction data, according to the signal configuration of described standard decoding the signal of high-quality decoding,
Wherein formed the difference signal between the signal of described source signal and described high-quality decoding, and described difference signal carried out lossless coding,
Wherein the difference signal of described lossless coding is packed with described frequency spectrum albefaction data and flows to form described lossless extension data,
Described method comprises step:
-described lossless extension data stream is unpacked so that described frequency spectrum albefaction data to be provided;
-data stream of described lossy coding is diminished decoding, thus the signal of described standard decoding rebuild, and use described frequency spectrum albefaction data, rebuild the signal of described high-quality decoding according to the signal of described standard decoding.
On the principle, another kind of decoding device of the present invention is suitable for the source signal data stream of lossless coding is decoded, described data stream is to obtain from the data stream of lossy coding and lossless extension data stream, the data stream of described lossy coding and lossless extension data stream form the data stream of the lossless coding of described source signal together
Wherein described source signal has been carried out lossy coding, described lossy coding provides the data stream and the frequency spectrum albefaction data of described lossy coding,
Wherein the data of described lossy coding have been carried out diminishing decoding accordingly, thereby have rebuild the signal of standard decoding, and used described frequency spectrum albefaction data, according to the signal configuration of described standard decoding the signal of high-quality decoding,
Wherein formed the difference signal between the signal of described source signal and described high-quality decoding, and described difference signal carried out lossless coding,
Wherein the difference signal of described lossless coding is packed with described frequency spectrum albefaction data and flows to form described lossless extension data,
Described equipment comprises:
-be suitable for described lossless extension data stream is unpacked so that the device of described frequency spectrum albefaction data to be provided;
-be suitable for the data stream of described lossy coding is diminished decoding, thus the signal of described standard decoding rebuild, and use described frequency spectrum albefaction data, rebuild the device of the signal of described high-quality decoding according to the signal of described standard decoding.
Favourable additional embodiment of the present invention is disclosed in each dependent claims.
Description of drawings
Example embodiment of the present invention has been described with reference to the drawings, in the accompanying drawing:
Fig. 1 shows the existing principle of lossless audio signal compression;
That Fig. 2 shows is existing, based on the block scheme of lossless encoder that diminishes and demoder;
Fig. 3 shows lossy encoder and diminishes the existing operate of demoder;
That Fig. 4 shows is of the present invention, based on the block scheme of the lossless coding that diminishes;
That Fig. 5 shows is of the present invention, based on the block scheme of the losslessly encoding that diminishes;
Fig. 6 shows the more detailed block scheme of the lossy encoder among Fig. 4;
Fig. 7 shows example signal:
A) discrete signal spectrum in the lossy encoder subband domain,
B) error signal after perception control quantizes,
C) error signal after the albefaction,
D) the pectrum noise shaping is with adaptive given LPC filter signal;
Fig. 8 shows the more detailed block scheme that diminishes demoder among Fig. 5;
Fig. 9 shows the lossless encoder among Fig. 4 and the more detailed block scheme of packing device;
Figure 10 shows the LPC decorrelator;
Figure 11 shows the harmless de-packetizer among Fig. 5 and the more detailed block scheme of demoder;
Figure 12 shows the extendfile structure;
Figure 13 shows bitstream formatization.
Embodiment
The present invention has solved the problem based on the suboptimum decorrelation of the lossless coding that diminishes by using improved lossy encoder (scrambler 41 as shown in Figure 4).Except according to original input signal S
PCMGenerate beyond (compliant) lossy bit stream 411 that adapts to, this scrambler also produces special frequency spectrum albefaction data, except out of Memory, described frequency spectrum albefaction data are sent to corresponding improved lossless encoder and the packing device 45 that diminishes demoder 42 and be sent to output lossless extension bit stream as supplementary 412.In Fig. 6, illustrate in greater detail lossy encoder 41.According in conjunction with what Fig. 6 and Fig. 7 explained, form frequency spectrum albefaction data.Diminish in the demoder 42 improved, lossy bit stream 44 is decoded and recovered the frequency spectrum of input signal present frame, thus the frequency spectrum albefaction data from signal 412 are added in the frequency spectrum.After this, in demoder 42, use the composite filter group, in subtracter 44, pass through from input signal S
PCMDeduct the signal S ' of respective decoder 42 outputs
DecCalculate time domain error signal S
Diff, wherein utilized 43 couples of described input signal S of impact damper
PCMCarried out corresponding delay, so that the needed processing time in scrambler 41 and the demoder 42 is compensated.Present error signal S
DiffHave white (just smooth (flat)) frequency power spectrum, this equates and have high decorrelation, thereby be suitable for entropy coding efficiently.With signal S
DiffBe fed to lossless encoder and packing device 45, wherein said lossless encoder and packing device 45 comprise entropy coder, and the lossy encoder side information data 412 that provides from scrambler 41 are provided in 451 outputs of the lossless extension of described lossless encoder and packing device 45 stream and diminish demoder side information data 421 from what demoder 42 provided.
In order to improve lossless coding efficient, improved lossy encoder 41 can reduce the amount (thereby reducing relevant bit rate) that (infavour of) is placed on the albefaction data of the additional LPC wave filter in lossless encoder and the packing device 45 of paying.Use lossy bit stream element (as zoom factor (scale factor) or the piece frequency spectrum (blockspectrum) in the demoder 42 in a preferred embodiment) to determine the LPC filter coefficient, only need to transmit very a spot of additional data so that can be at the decoder-side calculating filter coefficient.
In Fig. 5 based in the losslessly encoding that diminishes, improved diminishing in the demoder 51 lossy bit stream 411 is decoded, diminish the output signal S that demoder 51 output (known) diminishes decoding and coding
Dec(for example can be expressed as the mp3 signal of the decoding of the pattern of diminishing 1).
For example, flowed 451 o'clock, can in harmless de-packetizer and demoder 52, carry out consistance (consistency) check of coupling lossy bit stream 411 and the permission check that permission is decoded at different mode in the reception lossless extension.Different mode can be to diminish pattern 1, diminish pattern 2 and diminish mode 3.
If not only with pattern 1 operation, then in device 52, the frequency spectrum whitened signal that receives unpacked and (in out of Memory) is sent to as supplementary 521 and diminishes demoder 51, described diminishing in the demoder 51 frequency spectrum albefaction data are added in the frequency spectrum of recovery, and use the composite filter group to create output signal S '
DecIn diminishing pattern 2, S '
DecIt is output signal.This is to be better than signal S aspect perceived quality
DecDiminish signal, be called " intermediate mass " in the following description.Not necessarily will be to the difference signal S of lossless coding
DiffDecode.
In lossless mode 3, in device 52, lossless extension stream 451 is further unpacked and uses therein the entropy decoding, if it is synthetic to use optional LPC under the situation about correspondingly having represented in lossless extension bit stream 451.In a preferred embodiment, use is determined LPC composite filter coefficient from the corresponding information item (as the frequency spectrum of zoom factor in the subband domain that diminishes demoder 51 or relevant lossy coefficient block) of the data element of lossy bit stream 411 and the optional help information item of transmission in lossless extension stream 451.In device 52, recover error signal S
DiffAnd described S
DiffWith signal S '
DecSynchronously.Combined error signal S in totalizer
DiffWith signal S '
Dec, so that regain to original signal S
PCMMathematics on harmless reconstruction.
To signal S '
DecThe calculating aspect, diminish demoder 51 as operating the demoder 42 diminishing in the coded portion.Signal S ' in the decoded portion
DecWith the signal S ' in the coded portion
DecIdentical on the mathematics, and the signal S in the decoded portion
DiffWith the S in the coded portion
DiffIdentical on the mathematics.
Advantageously, can use integer arithmetic will diminish demoder realize 51 and 42 and install 52 and device 54 in optional LPC element be embodied as platform independence.
In conjunction with Fig. 6 lossy encoder 41 of key drawing 4 in more detail.In conjunction with Fig. 8 in more detail key drawing 5 diminish demoder 51.By the lossy encoder among Fig. 4 41 is made up with diminishing demoder 42, simplification is feasible.
Lossy encoder
In Fig. 7 a, described the example of the frequency spectrum of signal 611, shown the amplitude A of the frequency spectrum relative with frequency f.
According to the control of the sensor model that provides by counter 64, in first quantizer 62, signal 611 is quantized.Come error signal 651 by the sub-band sample 621 that from original sub-band sampling 611, deducts after the quantification.Usually the amplitude of this error signal is proportional with the masking threshold of determining in sensor model.Control signal 611 has been described example error signal 651 in Fig. 7 b.
In second quantization device 66, error signal 651 is quantized, make the other error signal 681 of calculating in the adaptive control loop that forms by second subtracter 68 and adaptation controller 67, wherein, as describing with signal 611 and 651 among Fig. 7 c, other error signal 681 is output signal poor of the signal 651 and second quantizer 66, and near white frequency spectrum.The output signal of second quantizer 66 is represented frequency spectrum albefaction data 661, and a part that is used as supplementary 412 is sent to and diminishes demoder 42 and be sent to lossless encoder and packing device 45.Adaptive control 67 controls second quantizer 66, and pay close attention to signal 661 and find correct quantification and correct bit rate.If thereby bit rate has surpassed predetermined threshold value and error spectrum 681 has not been estimated as " white ", then send (escape) signal 671 of escaping in supplementary 412, indication lossless encoder and packing device 45 should use additional LPC decorrelation.Adaptive control 67 is provided with the optimal quantization step-length at quantizer 66, to produce smooth noise floor (flat noise floor), referring to the signal among Fig. 7 c 681.This control can comprise the power analysis to signal 651.Iterative process is unnecessary.
Second task of adaptive control 67 is observation estimations to the bit rate of the signal behind the entropy coding 661.In step (step) or in the stage (stage) 93 signal 661 is carried out entropy coding subsequently.The bit rate of the signal 661 behind the entropy coding is the main contributions to the global rate of " can't harm " bit stream 451.If this bit rate is estimated to have surpassed threshold value, then be sent in the escape signal 671 that time domain is used additional LPC decorrelation.
In another embodiment, adaptive control 67 can make that signal 681 no longer is white (different quantization steps is used in just adaptive control 67 on frequency of b in axle) with signal 661 optimizations.Form the characteristic of noise floor 681 then with the given LPC decorrelator wave filter of coupling in the dictionary of Different L PC wave filter.Adaptive then control procedure becomes iteration, finds the most approaching coupling to signal 681 (just total bit rate) with least cost.This embodiment has been described in Fig. 7 d.
Diminish demoder
The demoder 42 that diminishes shown in Figure 8 receives lossy bit stream 441, in bit stream de-packetizer 81, described lossy bit stream 411 is unpacked, and decoding in sub-band sample demoder 82 (if be suitable for then comprise that inverse quantization device zoom factor handles), with produce with Fig. 6 in lossy encoder in the identical sub-band sample signal 821 of signal 621.In composite filter group 83 time domain is returned in signal 821 conversion, wherein said composite filter group 83 is restoring signal S in all cases
DecOne group of data value.In combiner 84, will (from lossless extension stream, receive) frequency spectrum albefaction data 661 after unpacking and add in the signal 821 forming signal 841, described signal 841 in subband domain, have with Fig. 6 and Fig. 7 c in the identical quantization error of quantization error of signal 681.Composite filter group 85 returns signal 841 conversion to time domain, and restoring signal S ' in all cases
DecOne group of data value.Because common output signal S
DecOr S '
DecSo, can use the single composite filter that is connected with signal 821 or signal 841 respectively.
Should use the operation of special integer arithmetic, realize diminishing demoder in the mode of platform independence.In the non-damage decoder of coding or decoding side is signal S ' with given bit stream decoding
DecNeed such as based on ARM, based on intel pentium or based on each platform the platform of DSP on generate identical result on the numeral.
Impact damper and synchronous
Lossy coding and decoding are included in the signal S of Fig. 4
PCMWith S '
DecBetween delay.When operation lossless encoder in stream is used in real time, lossy encoder is known this delay and will be controlled the first-in first-out in the impact damper 43 (First-In First-Out) buffering, guarantees accurate (just synchronous) operation of sampling with subtracter 44 places at Fig. 4.
When at file (for example being the file of lossless coding) when lossless encoder is operated in file operation, can replace impact damper 43 as the synchronous device of describing among the US-B-6903664 by using with the pcm audio file conversion.
In a preferred embodiment, lossy encoder will be inserted into the item of information of indication coding delay and source document length in the auxiliary data part of lossy bit stream of previous or two audio frames, and is inserted in first frame of lossless extension.Lossy demoder 42 and 51 will read this information and skip (zero) sampling of the coding of the front of being indicated by deferred message.
Lossless encoder and packing device
In Fig. 9, illustrate in greater detail lossless encoder and the packing device 45 of Fig. 4.During routine operation with error signal S
DiffThe height decorrelation, and can be to described error signal S in entropy coder 93
DiffEntropy coding, preferred embodiment uses the Golomb-Rice coding for this coding.Also use different entropy coding methods (for example Huffman coding) in scrambler 93, (from bus 412) special albefaction data 661 to be carried out entropy coding.Packing device 94 uses data 931 and the additional information item 412 (Tathagata is from the escape signal 671 that diminishes scrambler 41) behind the entropy coding to form the bit stream based on frame, and output lossless extension stream 451.Indicate if having the lossy encoder 41 of escape signal 671, then can in LPC decorrelator 91, use linear prediction error signal S
DiffFurther decorrelation illustrates in greater detail LPC and removes demoder 91 in Figure 10.LPC goes demoder 91 to receive help information from bus 421.Carry out switching by switch 92 according to (from bus 412) escape signal 671.
LPC removes demoder
LPC in Figure 10 goes in the demoder, in subtracter 101 from input signal S
DiffIn deduct version by fallout predictor 102.The output signal of subtracter is fed to switch 92.Fallout predictor 102 uses the wave filter that calculates by wave filter determiner (determinator) 103, and the filter coefficient of described wave filter is to obtain from the help information signal from bus 421.
Pattern 1
The zoom factor of demoder is transferred to wave filter determiner 103 as signal 421.Use these zoom factors si that the spectrum power of the remnants in the following transform domain is estimated: S
Ee(i)=2
-3/8si, i=0 wherein ..., N
Band-1, (number of bin), wherein power step by step estimates it can is level and smooth.
Duplicate these spectrum power values to form even number sequence S '
Ee(i), i=0 wherein ..., N
Band-1 ..., 2N
Band-1.Carry out this step to produce real-valued contrary FFT sequence.
After this, by iFFT (S '
Ee(i)) calculate auto-correlation.Can use Levison-Durbin to determine the LPC coefficient.
This process also can be used in the non-damage decoder.If in lossy encoder bit stream 411, do not transmit the more relevant portion of high frequency spectrum, then should the step/phase 63 of disappearance information 631 from lossy encoder be sent to packing device 94 transmitting, and be sent to wave filter determiner 103 from de-packetizer 111.
Pattern 2
Utilize adaptation controller 67 from the dictionary of LPS groups of filter coefficients, to select one group of LPC filter coefficient.Signal 631 becomes the dictionary index at selected group of coefficient then, and is passed to packing device 94 to transmit.
Supplementary
-(via bus 412 from scrambler 41 be sent to demoder 42 and to scrambler/packing device 45) frequency spectrum albefaction data 661 behind the coding;
(it is effective just indicating LPC decorrelation and LPC synthetic) escape signal 671 of the additional LPC decorrelation of-(being sent to scrambler/packing device 45 from scrambler 41) indication via bus 412;
-(being sent to scrambler/ packing device 45 or 94 respectively from demoder 42 via bus 421) help information signal just is used for the zoom factor that the LPC wave filter is determined;
-from lossy encoder 41 be sent to scrambler packing device 45/94, be used for help information 631, or the index of one group of predefined LPC filter coefficient at the zoom factor of high frequency band transmission deletion;
-at file to file applications, also have (be sent to scrambler 42 and be sent to scrambler/packing device 45 from scrambler 41) lossy encoder length of delay and/or source document length value via bus 421.
Based on the losslessly encoding that diminishes
As describing in conjunction with Fig. 2, use diminishes demoder 25 and non-damage decoder 27 is implemented decoding, wherein diminishes the output signal combination of demoder 25 and non-damage decoder 27 to regain original input signal sampling S with described
PCMAdvantageously, can implement decoding with different patterns.
Pattern 1
Demoder can be decoded and lossless extension not occur and be flowed 451 (compliant) lossy bit stream 411 of any adaptation, and signal S is provided
DecDo not provide lossless extension stream 451 occurring under the situation of permission of another pattern of use, this pattern is also effective.Preferably, demoder will be checked lossless extension stream, to mate the license ID in its rights database.
Pattern 2
When checking the lossless extension flow data, the permission check in the demoder also can start the intermediate mass pattern.Diminish demoder and only albefaction data 661 are used and unpacked, so that signal S ' to be provided
Dec
Mode 3
Can begin the lossless mode decoding after the sure permission assay, and output signal S
PCM
In Fig. 8, described in more detail and diminished demoder 51 accordingly.In from the supplementary that diminishes de-packetizer and demoder 52, informed operator scheme.Basically, suitable equally at diminishing demoder 42 described same detail.Lossy bit stream 411 behind the coding enters and is used for device 81 that bit stream is unpacked, follow thereafter be to be used for device 82 and composite filter group 83 that sub-band sample is decoded, diminish PCM signal S behind composite filter group 83 output decoders
DecWill be in totalizer 84 from the output signal and 661 combinations of corresponding frequency spectrum albefaction data of device 82.Signal 841 after the combination enters and diminishes PCM signal S ' behind the second composite filter group, 85, the second composite filter groups, 85 output decoders
Dec
Harmless de-packetizer and demoder
Figure 11 illustrates in greater detail harmless de-packetizer and demoder 52.Harmless de-packetizer 111 receives after the parsing and the stream of the lossless extension after unpacking 451.
Control information is routed to operation control 115, in described operation control 115, to the situation of file applications, can carry out consistency check with the integrality of identification about lossy bit stream 411 at file.As option, from lossless extension stream 451, extract reference fingerprint (for example CRC data) and the certain data block of lossy bit stream 411 is calculated current fingerprint.If these two fingerprints are identical, then proceed normal running.Can be used as the check of next step execute permission, with the operator scheme of identification permission.The corresponding information item 1151 that receives from external data base is used for comparing with the permission flag symbol of the bit stream that receives.Determine present mode, and corresponding signal 1152 is used to utilize supplementary passage 521 to be sent to for information about to diminish demoder 51.In specific embodiment, can also use to be used for device that the lossless extension stream after encrypting is deciphered.
After unpacking, in entropy decoder 112, audio frequency extension signal data 1111 is carried out the entropy decoding.For example, in demoder 112, the frequency spectrum albefaction data item behind the entropy coding is carried out corresponding entropy decoding.Decoded albefaction data 661 are sent to diminish demoder 51, with difference signal data S
DiffBe sent to combiner or sum unit 53.If identify the synthetic escape information of the additional LPC of application by de-packetizer 111 and operation control 115 in bit stream 451, described controller will use signal 1153 that switch 113 is switched to the LPC synthesis path.Use help information 1141 to calculate the coefficient of LPC composite filter 114, wherein said help information 1141 provides from de-packetizer 111, or can determine described help information 1141 according to the lossy bit stream zoom factor or according to the additional information of demoder subband signal 841 and transmission in lossless extension bit stream 451 (, or pointing to the index value of one group of predefined LPC coefficient) as the not zoom factor of disappearance of transmission or the spectrum power information of high frequency band in lossy bit stream 441.
Supplementary
Comprise following information and data element in the supplementary 521 that diminishes exchange between demoder 51 and harmless de-packetizer and the demoder 52:
-(being sent to demoder 52) mode indicators signal 1152;
-(being sent to demoder 52) frequency spectrum albefaction data 661;
-(being sent to harmless de-packetizer and demoder 52) is from diminishing demoder 42 to determine the help information of LPC filter coefficient;
-(being sent to demoder 52) is at lossy encoder length of delay and/or the source document length value of file to file applications.
The lossless extension bit stream
Can be used as a data element following data element be provided in the lossless extension bit stream:
-clearly discern the fingerprint of corresponding lossy bit stream.For 2 file applications, especially need this element, for container (file) and stream application, can ignore this element;
-mode indicators and corresponding D RM information;
-synchronizing information (lossy coding delay, source document length, end of file indicator);
The pcm word of-original signal long (16,20 or 24 bit);
-prompting point (cue point) information, described prompting dot information can carry out addressing to the lossless data frame quickly in (variable bit rate) stream inside, comprises the form of constant frame period pointer and frame period length indicator.
To file applications, only need provide once these items of information at file at the place that begins of harmless bit stream.In stream is used, need be every the out of Memory item of N frame transmission except that the prompting point data.
The frame data element of lossless extension bit stream is:
-make and can carry out the frame boundary indicator that frame synchronization is operated at lossy bit stream;
Error of spectrum (albefaction just) data of-coding;
The synthetic escape information of additional LPC is used in-indication, and the LPC help information;
The time error signal data of-coding.
Figure 12 illustrates the lossless extension stream file format.File header provides supplementary with the beginning decode procedure.Data from after, arranged to comprise the Frame of length available that is used to rebuild the intermediate mass sound signal and is used to rebuild the data of harmless quality audio signal.
The file header data:
-ID;
-head length degree;
-fingerprint (for example CRC32 data);
-pattern indication information piece;
-supplementary: codec delay, source document length, long, the sampling rate of pcm word;
-prompting point list data piece: the interval information in block length value, the frame, the number of table entries, pointer table.
Frame data:
-synchronization character (optionally) and frame length;
Error of spectrum (albefaction just) data of-coding: block length, coded data.
This is to be encoded to the needed data of intermediate mass (pattern 2).Under the situation that these data occur, will skip all the other frame data at the demoder that pattern is operated for 2 times;
-LPC help information: block length value, LPC mode indicators, coded data;
The time error signal of-coding: block length value, coded data.
Bitstream format and rights management
Referring to Figure 13, can should be used for formaing lossy bit stream 411 and lossless extension stream 451 at different storages or stream.To be fed to bitstream format device (formatter) 132 based on the output signal 411 and 451 of the lossless coding 131 that diminishes.The output signal 1322 that obtains can be single stream or file, maybe can comprise 2 streams or 2 files.Can come application rights management to handle by corresponding rights management data 1321 is provided for formatter 132.
In the decoding side, can use corresponding bit diffluence formatter.
Claims (20)
1. one kind is used the data stream (411) of lossy coding and lossless extension data to flow (451) to source signal (S
PCM) carrying out the method for lossless coding, data stream of described lossy coding (411) and lossless extension data stream (451) form the data stream (1322) of the lossless coding of described source signal together, said method comprising the steps of:
-described source signal is carried out lossy coding (41), wherein said lossy coding provides the data stream (411) and the frequency spectrum albefaction data (661) of described lossy coding;
-data of described lossy coding are diminished decoding (42) accordingly, thus the signal (S of reconstruction standard decoding
Dec), and use described frequency spectrum albefaction data, according to the signal of described standard decoding construct the high-quality decoding signal (S '
Dec);
-formation (43,44) described source signal (S
PCM) with the signal of described high-quality decoding (S '
Dec) between difference signal (S
Diff), and described difference signal carried out lossless coding (45);
-difference signal of described coding is packed with described frequency spectrum albefaction data (661), flow (451) to form described lossless extension data,
Wherein, described frequency spectrum albefaction data (661) produce in the following manner:
-in analysis filterbank (61) to described source signal (S
PCM) handle, and the output signal of described analysis filterbank (61) quantized (62), form output signal (611) of (65) analysis filterbank and the difference signal (651) between the quantized output signal (621), wherein said quantification is controlled by sensor model counter (64);
-output signal (611) and the described difference signal (651) between the quantized output signal (621) of analysis filterbank quantized (66), this further quantizes control (67), the input and the approaching white frequency spectrum of the difference signal (681) between the output that make described further quantification, thus the output signal of described further quantification forms described frequency spectrum albefaction data (661);
Wherein, described further quantification (66) is controlled by adaptation controller (67), described adaptation controller (67) is tested to the current bit rate of described frequency spectrum albefaction data (661), if described current bit rate surpasses predetermined threshold value, escape signal (671) is set then;
-wherein, to described difference signal (S
Diff) described lossless coding use entropy coder (93), only under the situation that is provided with described escape signal (671) input signal of described entropy coder (93) by LPC decorrelator (91).
2. according to the process of claim 1 wherein, described frequency spectrum albefaction data (661) are carried out entropy coding (93).
3. according to the process of claim 1 wherein, described LPC decorrelator is the LPC wave filter, uses item of information and/or help information item to determine the filter coefficient of described LPC wave filter.
4. according to the method for claim 3, wherein, described item of information is the frequency spectrum of the relevant coefficient block in the subband domain of data stream (411) of zoom factor and/or described lossy coding.
5. according to the process of claim 1 wherein, described lossless extension data stream (451) comprising:
The frequency spectrum albefaction data (661) of-coding;
-indication LPC decorrelation is the signal (671) of effectively escaping;
-help information signal;
-for lossy encoder length of delay and/or the source document length value of file to file applications.
6. source signal (S to lossless coding
PCM) the data stream method of decoding, described data stream is that data stream (411) and the lossless extension data stream (451) from lossy coding obtains, data stream of described lossy coding (411) and lossless extension data stream (451) form the data stream (1322) of the lossless coding of described source signal together
Wherein described source signal has been carried out lossy coding (41), described lossy coding provides the data stream (411) and the frequency spectrum albefaction data (661) of described lossy coding,
Wherein the data of described lossy coding have been carried out diminishing accordingly decoding (42), thus the signal (S of reconstruction standard decoding
Dec), and use described frequency spectrum albefaction data, according to the signal configuration of described standard decoding the signal of high-quality decoding (S '
Dec),
Wherein formed (43,44) described source signal (S
PCM) with the signal of described high-quality decoding (S '
Dec) between difference signal (S
Diff), and described difference signal carried out lossless coding (45),
Wherein the difference signal of described lossless coding is packed with described frequency spectrum albefaction data (661) and flows (451) to form described lossless extension data,
Wherein, described frequency spectrum albefaction data (661) produce in the following manner:
-in analysis filterbank (61) to described source signal (S
PCM) handle, and the output signal of described analysis filterbank (61) quantized (62), form output signal (611) of (65) analysis filterbank and the difference signal (651) between the quantized output signal (621), wherein said quantification is controlled by sensor model counter (64);
-output signal (611) and the described difference signal (651) between the quantized output signal (621) of analysis filterbank quantized (66), this further quantizes control (67), the input and the approaching white frequency spectrum of the difference signal (681) between the output that make described further quantification, thus the output signal of described further quantification forms described frequency spectrum albefaction data (661);
Wherein, described further quantification (66) is controlled by adaptation controller (67), described adaptation controller (67) is tested to the current bit rate of described frequency spectrum albefaction data (661), if described current bit rate surpasses predetermined threshold value, escape signal (671) is set then;
Described coding/decoding method comprises step:
-described lossless extension data stream (451) is unpacked (52) and the difference signal of described lossless coding is decoded (52), so that the difference signal (S of described lossless coding to be provided
Diff) and described frequency spectrum albefaction data (661), wherein, to the difference signal (S of described lossless coding
Diff) described decoding use entropy decoder (112), only be provided with the output signal of described entropy decoder (112) under the situation of described escape signal by LPC synthetic (114) at the coding place;
-data stream (411) of described lossy coding is diminished decoding (51), thus the signal (S of described standard decoding rebuild
Dec), and use described frequency spectrum albefaction data, according to the signal of described standard decoding rebuild (51) described high-quality decoding signal (S '
Dec);
-according to the difference signal (S of the lossless coding of described decoding
Diff) and according to the signal of described high-quality decoding (S '
Dec), form the source signal (S that (53) rebuild
PCM).
7. according to the method for claim 6, wherein, described frequency spectrum albefaction data (661) are carried out entropy coding (93), and described frequency spectrum albefaction data (661) are carried out entropy decoding (112) in decoding place at the coding place.
8. according to the method for claim 6, wherein, described LPC is synthetic to be carried out by the LPC wave filter, uses item of information and/or help information item to determine the filter coefficient of described LPC wave filter.
9. method according to Claim 8, wherein, described item of information is the frequency spectrum of the relevant coefficient block in the subband domain of data stream (411) of zoom factor and/or described lossy coding.
10. according to the method for claim 6, wherein, described lossless extension data stream (451) comprising:
The frequency spectrum albefaction data (661) of-coding;
-indication LPC decorrelation is the signal (671) of effectively escaping;
-help information signal;
-for lossy encoder length of delay and/or the source document length value of file to file applications.
11. one kind is used the data stream (411) of lossy coding and lossless extension data to flow (451) to source signal (S
PCM) carrying out the equipment of lossless coding, data stream of described lossy coding (411) and lossless extension data stream (451) form the data stream (1322) of the lossless coding of described source signal together, and described equipment comprises:
-being suitable for described source signal is carried out the device (41) of lossy coding, wherein said lossy coding provides the data stream (441) and the frequency spectrum albefaction data (661) of described lossy coding;
-be suitable for the data of described lossy coding are diminished decoding accordingly, thus the signal (S of reconstruction standard decoding
Dec), and use described frequency spectrum albefaction data, according to the signal of described standard decoding construct the high-quality decoding signal (S '
Dec) device (42);
-be suitable for forming described source signal (S
PCM) with the signal of described high-quality decoding (S '
Dec) between difference signal (S
Diff), described difference signal is carried out lossless coding, and the difference signal of described coding is packed forming the device (43,44,45) of described lossless extension data stream (451) with described frequency spectrum albefaction data (661),
Wherein, described lossy coding device comprises:
-be suitable in analysis filterbank described source signal (S
PCM) handle, and the output signal of described analysis filterbank quantized, form the output signal (611) of (65) analysis filterbank and the device (61 of the difference signal (651) between the quantized output signal (621), 62), wherein said quantification is controlled by sensor model counter (64);
-be suitable for the output signal (611) of analysis filterbank and the device (66) of the quantification of the described difference signal (651) between the quantized output signal (621), control should further quantize, the input and the approaching white frequency spectrum of the difference signal (681) between the output that make described further quantification, thereby the output signal of described further quantification forms described frequency spectrum albefaction data (661)
Wherein, described further quantification (66) is controlled by adaptation controller (67), described adaptation controller (67) is tested to the current bit rate of described frequency spectrum albefaction data (661), if described current bit rate surpasses predetermined threshold value, escape signal (671) is set then;
-wherein, form in the device, the difference signal (S of described lossless coding at described difference signal
Diff) described lossless coding use entropy coder (93), only under the situation that is provided with described escape signal (671) input signal of described entropy coder (93) by LPC decorrelator (91).
12., wherein, described frequency spectrum albefaction data (661) are carried out entropy coding (93) according to the equipment of claim 11.
13. according to the equipment of claim 11, wherein, described LPC decorrelator is the LPC wave filter, uses item of information and/or help information item to determine the filter coefficient of described LPC wave filter.
14. according to the equipment of claim 13, wherein, described item of information is the frequency spectrum of the relevant coefficient block in the subband domain of data stream (411) of zoom factor and/or described lossy coding.
15. according to the equipment of claim 11, wherein, described lossless extension data stream (451) comprising:
The frequency spectrum albefaction data (661) of-coding;
-indication LPC decorrelation is the signal (671) of effectively escaping;
-help information signal;
-for lossy encoder length of delay and/or the source document length value of file to file applications.
16. source signal (S to lossless coding
PCM) the data stream equipment of decoding, described data stream is that data stream (411) and the lossless extension data stream (451) from lossy coding obtains, data stream of described lossy coding (411) and lossless extension data stream (451) form the data stream (1322) of the lossless coding of described source signal together
Wherein described source signal has been carried out lossy coding (41), described lossy coding provides the data stream (411) and the frequency spectrum albefaction data (661) of described lossy coding,
Wherein the data of described lossy coding are carried out diminishing accordingly decoding (42), thereby rebuild the signal (S of standard decoding
Dec), and use described frequency spectrum albefaction data, according to the signal configuration of described standard decoding the signal of high-quality decoding (S '
Dec),
Wherein formed (43,44) described source signal (S
PCM) with the signal of described high-quality decoding (S '
Dec) between difference signal (S
Diff), and described difference signal carried out lossless coding (45),
Wherein the difference signal of described lossless coding is packed with described frequency spectrum albefaction data (661) and flows (451) to form described lossless extension data,
Wherein, described frequency spectrum albefaction data (661) produce in the following manner:
-in analysis filterbank (61) to described source signal (S
PCM) handle, and the output signal of described analysis filterbank (61) quantized (62), form output signal (611) of (65) analysis filterbank and the difference signal (651) between the quantized output signal (621), wherein said quantification is controlled by sensor model counter (64);
-output signal (611) and the described difference signal (651) between the quantized output signal (621) of analysis filterbank quantized (66), this further quantizes control (67), the input and the approaching white frequency spectrum of the difference signal (681) between the output that make described further quantification, thereby the output signal of described further quantification forms described frequency spectrum albefaction data (661)
Wherein, described further quantification (66) is by adaptation controller (67) control, and described adaptation controller (67) is tested to the current bit rate of described frequency spectrum albefaction data (661), if described current bit rate surpasses predetermined threshold value, escape signal (671) then is set
Described decoding device comprises:
-be suitable for described lossless extension data stream (451) is unpacked and the difference signal of described lossless coding is decoded, so that the difference signal (S of described lossless coding to be provided
Diff) and the device (52) of described frequency spectrum albefaction data (661), wherein, to the difference signal (S of described lossless coding
Diff) described decoding use entropy decoder (112), only be provided with the output signal of described entropy decoder (112) under the situation of described escape signal by LPC synthetic (114) at the coding place;
-the data stream (411) that is suitable for described lossy coding diminishes decoding, thereby rebuilds the signal (S of described standard decoding
Dec), and use described frequency spectrum albefaction data, according to the signal of described standard decoding rebuild (51) described high-quality decoding signal (S '
Dec) device (51);
-be suitable for difference signal (S according to the lossless coding of described decoding
Diff) and according to the signal of described high-quality decoding (S '
Dec) form the source signal (S of reconstruction
PCM) device (53).
17. according to the equipment of claim 16, wherein, described frequency spectrum albefaction data (661) are carried out entropy coding (93), and described frequency spectrum albefaction data (661) are carried out entropy decoding (112) in decoding place at the coding place.
18. according to the equipment of claim 16, wherein, described LPC is synthetic to be carried out by the LPC wave filter, uses item of information and/or help information item to determine the filter coefficient of described LPC wave filter.
19. according to the equipment of claim 18, wherein, described item of information is the frequency spectrum of the relevant coefficient block in the subband domain of data stream (411) of zoom factor and/or described lossy coding.
20. according to the equipment of claim 16, wherein, described lossless extension data stream (451) comprising:
The frequency spectrum albefaction data (661) of-coding;
-indication LPC decorrelation is the signal (671) of effectively escaping;
-help information signal;
-for lossy encoder length of delay and/or the source document length value of file to file applications.
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| EP06113576.0 | 2006-05-05 | ||
| EP06113576A EP1852848A1 (en) | 2006-05-05 | 2006-05-05 | Method and apparatus for lossless encoding of a source signal using a lossy encoded data stream and a lossless extension data stream |
| PCT/EP2007/053783 WO2007128661A1 (en) | 2006-05-05 | 2007-04-18 | Method and apparatus for lossless encoding of a source signal using a lossy encoded data stream and a lossless extension data stream |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101432610A CN101432610A (en) | 2009-05-13 |
| CN101432610B true CN101432610B (en) | 2011-11-23 |
Family
ID=36694486
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2007800156126A Expired - Fee Related CN101432610B (en) | 2006-05-05 | 2007-04-18 | Method and apparatus for lossless encoding of a source signal using a lossy encoded data stream and a lossless extension data stream |
Country Status (9)
| Country | Link |
|---|---|
| US (1) | US8428941B2 (en) |
| EP (2) | EP1852848A1 (en) |
| JP (1) | JP5135330B2 (en) |
| KR (1) | KR101404335B1 (en) |
| CN (1) | CN101432610B (en) |
| AT (1) | ATE459868T1 (en) |
| BR (1) | BRPI0711190B1 (en) |
| DE (1) | DE602007005119D1 (en) |
| WO (1) | WO2007128661A1 (en) |
Families Citing this family (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1881485A1 (en) * | 2006-07-18 | 2008-01-23 | Deutsche Thomson-Brandt Gmbh | Audio bitstream data structure arrangement of a lossy encoded signal together with lossless encoded extension data for said signal |
| CN101615395B (en) | 2008-12-31 | 2011-01-12 | 华为技术有限公司 | Methods, devices and systems for encoding and decoding signals |
| US8457976B2 (en) | 2009-01-30 | 2013-06-04 | Qnx Software Systems Limited | Sub-band processing complexity reduction |
| EP2755205B1 (en) * | 2010-01-29 | 2019-12-11 | 2236008 Ontario Inc. | Sub-band processing complexity reduction |
| EP2395505A1 (en) * | 2010-06-11 | 2011-12-14 | Thomson Licensing | Method and apparatus for searching in a layered hierarchical bit stream followed by replay, said bit stream including a base layer and at least one enhancement layer |
| TWI716169B (en) * | 2010-12-03 | 2021-01-11 | 美商杜比實驗室特許公司 | Audio decoding device, audio decoding method, and audio encoding method |
| EP2676266B1 (en) | 2011-02-14 | 2015-03-11 | Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. | Linear prediction based coding scheme using spectral domain noise shaping |
| PT2676267T (en) | 2011-02-14 | 2017-09-26 | Fraunhofer Ges Forschung | Encoding and decoding of pulse positions of tracks of an audio signal |
| MY164797A (en) | 2011-02-14 | 2018-01-30 | Fraunhofer Ges Zur Foederung Der Angewandten Forschung E V | Apparatus and method for processing a decoded audio signal in a spectral domain |
| AU2012217216B2 (en) * | 2011-02-14 | 2015-09-17 | Fraunhofer-Gesellschaft Zur Foerderung Der Angewandten Forschung E.V. | Apparatus and method for coding a portion of an audio signal using a transient detection and a quality result |
| GB201210373D0 (en) * | 2012-06-12 | 2012-07-25 | Meridian Audio Ltd | Doubly compatible lossless audio sandwidth extension |
| KR102204136B1 (en) | 2012-08-22 | 2021-01-18 | 한국전자통신연구원 | Apparatus and method for encoding audio signal, apparatus and method for decoding audio signal |
| EP2770642B1 (en) * | 2013-02-26 | 2022-10-05 | Tata Consultancy Services Limited | Systems and methods for data archival |
| EP2830064A1 (en) * | 2013-07-22 | 2015-01-28 | Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. | Apparatus and method for decoding and encoding an audio signal using adaptive spectral tile selection |
| US9286313B1 (en) * | 2014-12-27 | 2016-03-15 | Ascava, Inc. | Efficient lossless reduction of data by deriving data from prime data elements resident in a content-associative sieve |
| US10431242B1 (en) * | 2017-11-02 | 2019-10-01 | Gopro, Inc. | Systems and methods for identifying speech based on spectral features |
| SG11202004430YA (en) * | 2017-11-17 | 2020-06-29 | Fraunhofer Ges Forschung | Apparatus and method for encoding or decoding directional audio coding parameters using different time/frequency resolutions |
| CN109887515B (en) * | 2019-01-29 | 2021-07-09 | 北京市商汤科技开发有限公司 | Audio processing method and device, electronic equipment and storage medium |
| CN115472171A (en) * | 2021-06-11 | 2022-12-13 | 华为技术有限公司 | Encoding and decoding method, apparatus, device, storage medium, and computer program |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1262815A (en) * | 1998-04-09 | 2000-08-09 | 皇家菲利浦电子有限公司 | Lossless encoding/decoding in transmission system |
Family Cites Families (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU704693B2 (en) | 1994-12-20 | 1999-04-29 | Dolby Laboratories Licensing Corporation | Method and apparatus for applying waveform prediction to subbands of a perceptual coding system |
| DE19742201C1 (en) * | 1997-09-24 | 1999-02-04 | Fraunhofer Ges Forschung | Method of encoding time discrete audio signals, esp. for studio use |
| JPH11109996A (en) * | 1997-10-06 | 1999-04-23 | Victor Co Of Japan Ltd | Voice coding device, voice coding method and optical recording medium recorded with voice coding information and voice decoding device |
| KR100743534B1 (en) | 2000-01-07 | 2007-07-27 | 코닌클리케 필립스 일렉트로닉스 엔.브이. | Transmission device and method for transmitting a digital information |
| SE0004163D0 (en) * | 2000-11-14 | 2000-11-14 | Coding Technologies Sweden Ab | Enhancing perceptual performance or high frequency reconstruction coding methods by adaptive filtering |
| US6675148B2 (en) | 2001-01-05 | 2004-01-06 | Digital Voice Systems, Inc. | Lossless audio coder |
| JP3636094B2 (en) * | 2001-05-07 | 2005-04-06 | ソニー株式会社 | Signal encoding apparatus and method, and signal decoding apparatus and method |
| KR100908114B1 (en) | 2002-03-09 | 2009-07-16 | 삼성전자주식회사 | Scalable lossless audio encoding / decoding apparatus and method thereof |
| US7275036B2 (en) | 2002-04-18 | 2007-09-25 | Fraunhofer-Gesellschaft Zur Foerderung Der Angewandten Forschung E.V. | Apparatus and method for coding a time-discrete audio signal to obtain coded audio data and for decoding coded audio data |
| US7536305B2 (en) | 2002-09-04 | 2009-05-19 | Microsoft Corporation | Mixed lossless audio compression |
| KR100467326B1 (en) * | 2002-12-09 | 2005-01-24 | 학교법인연세대학교 | Transmitter and receiver having for speech coding and decoding using additional bit allocation method |
| DE102004007184B3 (en) | 2004-02-13 | 2005-09-22 | Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. | Method and apparatus for quantizing an information signal |
| US7392195B2 (en) | 2004-03-25 | 2008-06-24 | Dts, Inc. | Lossless multi-channel audio codec |
| US20060015329A1 (en) | 2004-07-19 | 2006-01-19 | Chu Wai C | Apparatus and method for audio coding |
| CA2603246C (en) * | 2005-04-01 | 2012-07-17 | Qualcomm Incorporated | Systems, methods, and apparatus for anti-sparseness filtering |
| KR100818268B1 (en) | 2005-04-14 | 2008-04-02 | 삼성전자주식회사 | Apparatus and method for audio encoding/decoding with scalability |
| US8055500B2 (en) | 2005-10-12 | 2011-11-08 | Samsung Electronics Co., Ltd. | Method, medium, and apparatus encoding/decoding audio data with extension data |
| US7835904B2 (en) | 2006-03-03 | 2010-11-16 | Microsoft Corp. | Perceptual, scalable audio compression |
| US20080243518A1 (en) | 2006-11-16 | 2008-10-02 | Alexey Oraevsky | System And Method For Compressing And Reconstructing Audio Files |
| KR101379263B1 (en) | 2007-01-12 | 2014-03-28 | 삼성전자주식회사 | Method and apparatus for decoding bandwidth extension |
| MY146431A (en) | 2007-06-11 | 2012-08-15 | Fraunhofer Ges Forschung | Audio encoder for encoding an audio signal having an impulse-like portion and stationary portion, encoding methods, decoder, decoding method, and encoded audio signal |
-
2006
- 2006-05-05 EP EP06113576A patent/EP1852848A1/en not_active Withdrawn
-
2007
- 2007-04-18 EP EP07728245A patent/EP2016383B1/en not_active Not-in-force
- 2007-04-18 CN CN2007800156126A patent/CN101432610B/en not_active Expired - Fee Related
- 2007-04-18 BR BRPI0711190-8A patent/BRPI0711190B1/en not_active IP Right Cessation
- 2007-04-18 AT AT07728245T patent/ATE459868T1/en not_active IP Right Cessation
- 2007-04-18 DE DE602007005119T patent/DE602007005119D1/en active Active
- 2007-04-18 KR KR1020087027032A patent/KR101404335B1/en active IP Right Grant
- 2007-04-18 WO PCT/EP2007/053783 patent/WO2007128661A1/en active Application Filing
- 2007-04-18 JP JP2009508301A patent/JP5135330B2/en not_active Expired - Fee Related
- 2007-04-18 US US12/226,992 patent/US8428941B2/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1262815A (en) * | 1998-04-09 | 2000-08-09 | 皇家菲利浦电子有限公司 | Lossless encoding/decoding in transmission system |
Non-Patent Citations (2)
| Title |
|---|
| ADISTAMBHA K ET AL.An Investigation into Embedded Audio Coding using An AAC Perceptually Lossless Base Layer.《PROCEEDINGS OF THE AUSTRALIAN INTERNATIONAL CONFERENCE ON SPEECH SCIENCE AND TECHNOLOGY》.2004,227页右栏倒数最后一段至228页右栏60行,图1. * |
| JIN A ET AL.Scalable audio coder based on quantizer units of MDCT coefficients.《ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, 1999. PROCEEDINGS》.1999,898页左栏第一段至899页左栏第一段,图5. * |
Also Published As
| Publication number | Publication date |
|---|---|
| EP1852848A1 (en) | 2007-11-07 |
| EP2016383A1 (en) | 2009-01-21 |
| JP5135330B2 (en) | 2013-02-06 |
| DE602007005119D1 (en) | 2010-04-15 |
| CN101432610A (en) | 2009-05-13 |
| US8428941B2 (en) | 2013-04-23 |
| BRPI0711190A2 (en) | 2011-08-23 |
| JP2009536363A (en) | 2009-10-08 |
| ATE459868T1 (en) | 2010-03-15 |
| EP2016383B1 (en) | 2010-03-03 |
| KR20090007395A (en) | 2009-01-16 |
| BRPI0711190B1 (en) | 2018-01-30 |
| US20090164226A1 (en) | 2009-06-25 |
| KR101404335B1 (en) | 2014-06-09 |
| WO2007128661A1 (en) | 2007-11-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101432610B (en) | Method and apparatus for lossless encoding of a source signal using a lossy encoded data stream and a lossless extension data stream | |
| KR101664434B1 (en) | Method of coding/decoding audio signal and apparatus for enabling the method | |
| CN1878001B (en) | Apparatus and method of encoding audio data, and apparatus and method of decoding encoded audio data | |
| US7876966B2 (en) | Switching between coding schemes | |
| TWI498881B (en) | Improved decoding of multichannel audio encoded bit streams using adaptive hybrid transformation | |
| CN101490748B (en) | Method and apparatus for lossless encoding of a source signal, using a lossy encoded data stream and a lossless extension data stream | |
| CN101484937B (en) | Decoding of predictively coded data using buffer adaptation | |
| EP1749296A1 (en) | Multichannel audio extension | |
| CN101371296B (en) | Apparatus and method for encoding and decoding signal | |
| WO2002103685A1 (en) | Encoding apparatus and method, decoding apparatus and method, and program | |
| CN101432802A (en) | Method and apparatus for lossless encoding of a source signal, using a lossy encoded data stream and a lossless extension data stream | |
| CN102150202A (en) | Method and apparatus to encode and decode an audio/speech signal | |
| KR20100007738A (en) | Apparatus for encoding and decoding of integrated voice and music | |
| US20110238424A1 (en) | Method and apparatus for encoding and decoding excitation patterns from which the masking levels for an audio signal encoding and decoding are determined | |
| KR20100089772A (en) | Method of coding/decoding audio signal and apparatus for enabling the method | |
| KR102622804B1 (en) | Backward-compatible integration of harmonic transposer for high frequency reconstruction of audio signals | |
| JP2010506207A (en) | Encoding method, decoding method, encoder, decoder, and computer program product | |
| KR101387808B1 (en) | Apparatus for high quality multiple audio object coding and decoding using residual coding with variable bitrate | |
| CA2765523A1 (en) | Coding method, decoding method, and device and recording medium using the methods | |
| KR102546098B1 (en) | Apparatus and method for encoding / decoding audio based on block | |
| KR20080092823A (en) | Apparatus and method for encoding and decoding signal | |
| CN102568489B (en) | Scrambler | |
| KR101455648B1 (en) | Method and System to Encode/Decode Audio/Speech Signal for Supporting Interoperability | |
| CN101103531A (en) | Scalable audio data arithmetic decoding method, medium, and apparatus, and method, medium, and apparatus truncating audio data bitstream | |
| Gray | A Partial Hstry of Losy Compression. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20170522 Address after: Amsterdam, The Netherlands Patentee after: DOLBY INTERNATIONAL AB Address before: French Boulogne - Bilang Kurt Patentee before: THOMSON LICENSING |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20180628 Address after: No. 18, Wu Sha seashore road, Changan Town, Dongguan, Guangdong Patentee after: GUANGDONG OPPO MOBILE TELECOMMUNICATIONS Corp.,Ltd. Address before: Amsterdam, The Netherlands Patentee before: Dolby International AB |
|
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20111123 |